Bayesian Mars Interior Model Blind-Predicts Crater Locations Within ~342 km
DynamicWebPaige · x · 2026-11-21
- Blind validation: Google's DynamicWebPaige shared a Bayesian inversion study of Mars' interior. Six fresh craters photographed by NASA's orbiter (heard by InSight) were hidden from the model: for a 150 m crater 2,150 miles away, it predicted the location within 2.8° and 342 km.
- Core debate settled by data: one giant liquid iron core vs. a smaller core in a 107 km melted-rock blanket — echoes alone tie, so cosmochemistry decides; the melted-blanket model cut blind crater error by 30%.
- Other results: InSight heard 71× more quakes on quiet evenings (wind, not Phobos); zero aftershocks 48h after Mars' biggest quake.
- Method: 16-temperature parallel-tempering MCMC jointly fitting InSight seismic waveforms, NASA MRO craters and RISE ranging, with zero-leakage priors. Author notes such charts once took two all-nighters.
More from Research
- Lean community certifies Riemann proof; honeypot fake keys expose breaches at three AI providers — LeviTurk · 2026-11-22
- Riemann hypothesis 'proven by Turkish mathematician using AI' claim mocked as the 'Organic Turk' — CatAstro_Piyush · 2026-11-22
- Andriy Burkov releases The Hundred-Page Language Models Book, hands-on with PyTorch — burkov · 2026-11-22
- Generative AI meets cardiology: Dagstuhl seminar bridges ML and clinical practice — StephanMandt · 2026-11-22
- SubRoformer hits SOTA music source separation with DSP subband filters — affige_yang · 2026-11-22
- Emily Riehl releases Lean game teaching dependent type theory at intro-to-proofs level — abhi9u · 2026-11-22